THE CHALLENGE

ADS was looking for efficient, powerful cloud computing solutions for their weather modeling products. The agencies and companies ADS works with are often constrained within a limited budget for each project. The nature of weather modeling means that the faster the cloud HPC cluster processes data, the more models can be run each day, leading to more accurate and useful information. Therefore, ADS needed to find cloud HPC solutions that provided the most computing power within each budget range. In addition, custom software must be installed before the weather modeling solution can be delivered to clients, which necessitated ADS personnel work on-site with the cloud provider to ensure that the solutions were ready to use upon delivery to the client. In addition, because the engineers and scientists working with the models may not have additional expertise in implementing and supporting the cloud cluster when delivered, ADS required a turnkey cloud solution for its clients that was ready to use upon delivery.

THE SOLUTION

Scott Capps started working with NZO Cloud in 2011 while seeking a high-performance cloud computing cluster quote for SDG&E. NZO Cloud stood out among their competitors for many reasons. Most importantly, they took the time to really understand the cloud computing needs and provided solutions that met those needs while not trying to up-sell into excessive cloud hardware. Capps says, “When designing a system that will run weather models, it is easy to equip it with cloud hardware with capabilities beyond what is actually needed.” Capps continues, “As a result, customers pay much more for a cloud system and only utilize a fraction of its full potential.” In contrast, NZO Cloud worked with Capps to customize the cloud HPC while staying within the budget of his client.

In one of the most recent collaborations with ADS, NZO Cloud was able to deliver a custom high performance Cloud instance for an ADS client to run weather models. The cloud cluster included 560 Total Intel Xeon E5-2600v4 Process Cores, 1120 Cores with Hyperthreading Enabled, as well as 2112 GB High-Performance DDR ECC Registered System Memory, 3.7 GB per Processor Core. The instance included 94 TB Raw Storage Capacity and Mellanox InfiniBand Connect X-4 100 Gbps High-Performance Network Backplane with remote management network. The cluster ran on Red Hat Linux OS with NZO Cloud CBeST v 4.0 Cluster Management Toolkit and Rack & Roll Integration for Turn-Key Deployment, making it easily deployable upon delivery. Best of all, ADS Principal Scott Capps was able to work with NZO Cloud at their facilities to ensure that all custom software and toolkits unique to the specific weather modeling solution were pre-installed and optimized before the cloud cluster was delivered.

When designing a system that will run weather models, it is easy to equip it with cloud hardware with capabilities beyond what is actually needed. As a result, customers pay much more for a cloud system and only utilize a fraction of its full potential.

Scott Capps Atmospheric Data Solutions

Atmospheric Data Solutions, LLC (ADS) works with public and private agencies to develop atmospheric science products that help mitigate and manage risk from severe weather and future climate change. NZO Cloud works closely with ADS to create customized weather modeling solutions which require the design, implementation, and support of high-performance cloud instances. The weather modeling solutions that ADS creates include high impact weather forecast guidance products, tailored regional wildfire forecast guidance products, and utility load and outage forecasts, all requiring analysis of a large quantity of data that demands high-performance cloud computing to maximize accuracy and the number of times models can be run daily. The work ADS produces in partnership with utility companies and other public agencies benefits public welfare and safety.